agent-toolkit
A collection of project-agnostic skills for common engineering tasks such as ticket refinement, planning, code reviews, agent memory hygiene, skill authoring, and more.
Safety scan
A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.
Check it deeper on your machine
It has changed since its scan on 11 Nov 2026.
As a Claude Code plugin, install it and then ask Claude Code:
/plugin marketplace add keithmackay/mackayi/plugin install scanskillsafety@mackayiDo a deep scan of https://github.com/eai-org/agent-toolkit with scanskillsafety
Or from a terminal, clone it and run it:
git clone https://github.com/keithmackay/scanskillsafety-skill ~/.claude/skills/scanskillsafetynode ~/.claude/skills/scanskillsafety/cli.cjs --deep 'https://github.com/eai-org/agent-toolkit'
It checks the exact commit: every file, its packages against OSV (api.osv.dev), how it compares with this page's scan (findsafeskills.com receives the repository's address), and, if you agree, Claude reading every file with your own Claude subscription. It never runs the code.
Badge for your README
Markdown
[](/listings/eai-org--agent-toolkit)HTML
<a href="/listings/eai-org--agent-toolkit"><img src="/api/listings/eai-org--agent-toolkit/badge.svg" alt="findsafeskills safety rating" /></a>Install
Install instructions from the README
As written by the repo's author. findsafeskills has not verified these steps.
Install all skills in one command:
git clone https://github.com/eai-org/agent-toolkit.git && cd agent-toolkit && ./install.sh
Update in one command:
cd agent-toolkit && git pull && ./install.sh
On Claude Code the install also registers a hook that runs that update for you once a day; an install made before it needs the command above once to pick it up ([docs/auto-update.md](./docs/auto-update.md)).
Every skill is installed by default. To leave one out, name it once and the choice sticks, daily update included:
./install.sh --exclude use-conversational-language
`--include <name>` puts it back. Both are repeatable.
On Windows the skills link through junctions, so the update command above works as-is, though an install predating them needs one `./install.sh --force` to convert. The rules are copied instead and need `--force` every time (see [Windows](./docs/install-skills.md#windows)).
How the symlink install works and the other install methods — hand-picking skills, other agents, [skills.sh](https://skills.sh/), the Claude Code plugin marketplace — are covered in [docs/install-skills.md](./docs/install-skills.md).
What was scanned
The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.
.claude-plugin/marketplace.json · Manifest · 11 lines
View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)
This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.
- {
- "name": "agent-toolkit",
- "owner": {
- "name": "EAI",
- "email": "engineering.age.ai@gmail.com"
- },
- "description": "A collection of project-agnostic skills for common engineering tasks such as ticket refinement, planning, code reviews, agent memory hygiene, skill authoring, and more.",
- "plugins": [
- { "name": "agent-toolkit", "source": "./" }
- ]
- }
README.md · README · 360 lines
View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)
This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.
- # agent-toolkit
- A collection of generic agentic tools for common engineering tasks, designed to work with any AI
- agent on any kind of software project. The philosophy behind the toolkit — five pillars and a
- litmus test for every new artifact — is captured in
- [docs/core-philosophy.md](./docs/core-philosophy.md).
- ## Skills
- Agentic skills for everyday software engineering — solo or in a team, not bound to any specific
- language or framework.
- ### Skill & doc authoring
- Tools to create and continuously improve the skills and docs your agents rely on, following
- [this approach to agentic skills](https://medium.com/engineering-in-the-age-of-ai/my-approach-to-agentic-skills-e08dc6c0d1cd).
- - **[compact-docs-writer](./skills/compact-docs-writer/SKILL.md)** — write docs with maximum token
- economy.
- - **[compact-skill-creator](./skills/compact-skill-creator/SKILL.md)** — create or edit skills,
- keeping them lean and efficient.
- - **[self-improve](./skills/self-improve/SKILL.md)** — capture a lesson into the skill, doc or check
- that governs it, so mistakes aren't repeated and agents keep getting better at the project.
- ### Context & memory hygiene
- Maintenance to run from time to time, keeping your setup tidy and your context sharp.
- - **[agentify-project](./skills/agentify-project/SKILL.md)** — make a project agent-ready by
- converging it toward a canonical agent-neutral structure (a lean AGENTS.md index with
- progressive disclosure, shared skills, gitignore hygiene). Re-runnable, and doubles as an audit.
- - **[context-checkup](./skills/context-checkup/SKILL.md)** — audit what auto-loads into a
- session's context and spot what can be trimmed to reduce startup tokens.
- [Why this is important](https://medium.com/engineering-in-the-age-of-ai/keep-your-ai-agents-context-window-sharp-7255d83a8949).
- - **[memory-doctor](./skills/memory-doctor/SKILL.md)** — drain the memory your agents keep
- auto-accumulating, block by block, moving each entry to the right place or archiving it.
- [More about memory-doctor](https://medium.com/engineering-in-the-age-of-ai/keep-your-ai-agents-memory-clean-and-organized-with-memory-doctor-a79f7174f257).
- ### Task workflow
- A daily routine for any programming task, following the
- [RPAC workflow](https://medium.com/engineering-in-the-age-of-ai/the-refine-plan-act-pattern-for-agentic-ai-coding-59ee013e4427):
- fetch a ticket, refine it, harden it, plan it, harden it again, split the plan into small tasks,
- then let a fresh session execute them one at a time and consolidate the changes.
- - **[fetch-ticket](./skills/fetch-ticket/SKILL.md)** — download a ticket from any tracker
- (e.g. GitHub, Jira, Azure DevOps) and save it as a self-contained markdown file.
- - **[attach-to-ticket](./skills/attach-to-ticket/SKILL.md)** — save a pasted screenshot, a file
- or a URL beside a fetched ticket and reference it from the ticket file, e.g. to fill an
- attachment the fetch could not download.
- - **[grill-me](./skills/grill-me/SKILL.md)** — get interviewed relentlessly about a plan, design
- or idea, one question at a time, each with a recommended answer, until every branch of the
- decision tree is resolved; ends with a paste-ready list of the settled decisions. Adapted from
- [Matt Pocock's grill-me](https://github.com/mattpocock/skills) (MIT, copyright 2026 Matt Pocock).
- - **[refine-ticket](./skills/refine-ticket/SKILL.md)** — define the "what" of a task: validate the
- ticket — or a raw idea you want to brainstorm — against the codebase, settle open decisions
- together, and save a self-contained requirements doc a fresh session can pick up. Takes a
- fetched ticket file, or a ticket URL a subagent fetches first.
- - **[create-implementation-plan](./skills/create-implementation-plan/SKILL.md)** — define the
- "how" of a task: turn the requirements into an implementation plan, settling the technical
- decisions together, then save it for a fresh session to execute.
- - **[harden-artifact](./skills/harden-artifact/SKILL.md)** — harden a requirements doc or a plan
- before the next phase picks it up: a fresh-context reviewer tests every deviation from the
- ticket and every claim against the code and related tickets, the session challenges each
- finding, and proven mistakes get fixed in the document (say `autoaccept` to skip the asks for
- those). Findings that reveal a durable lesson end in an offer to run self-improve, when
- installed.
- - **[split-plan-tasks](./skills/split-plan-tasks/SKILL.md)** — break a plan into individually
- reviewable tasks grouped into PR-sized batches, appended to the plan file for a fresh session to
- execute one at a time.
- [Why not execute the full plan at once](https://medium.com/engineering-in-the-age-of-ai/do-not-execute-the-full-plan-break-it-down-step-by-step-2e0ba30ddf73).
- - **[execute-plan-tasks](./skills/execute-plan-tasks/SKILL.md)** — execute one task of that
- breakdown at a time: implement it, verify it, tick it off in the plan, and hand it back for
- review before the next; when a group is done, hand it to self-review (if installed) before the
- next group starts. A task whose approach fails twice, or needs reshaping, stops unticked with a
- hand-off to re-plan or harden the plan (if installed).
- - **[create-manual-test-instructions](./skills/create-manual-test-instructions/SKILL.md)** —
- derive manual test steps from a ticket or requirements file, or from a code change (branch,
- PR, commit range, diff), useful for the developer or QA.
- - **[handover](./skills/handover/SKILL.md)** — package a finished change for its reviewers:
- what it does and why, the decisions and plan deviations worth knowing, a review guide, and a
- diagram or tree wherever the diff alone hides the change's shape — paste-ready as the PR
- description. [More about handover](https://medium.com/engineering-in-the-age-of-ai/the-missing-step-in-agentic-coding-the-handover-d1963c3d2c1d).
- - **[prepare-prompt](./skills/prepare-prompt/SKILL.md)** — write the prompt that hands this
- session's work to a fresh session — what exists, what is open, what to do and in what form, the
- facts it cannot derive — as short as the next session allows.
- ### Review assistants
- Review helpers that check the codebase while assisting with code or ticket reviews.
- - **[fetch-pr-review](./skills/fetch-pr-review/SKILL.md)** — collect the comments left by other
- reviewers on your PR and save them into a markdown doc, pinned to the PR head, ready to address
- (or push back on), for example via refine-pr-review.
- - **[refine-pr-review](./skills/refine-pr-review/SKILL.md)** — go through a fetched PR review
- together, comment by comment — address, partial, or push back — drafting the replies, turning
- the accepted changes into a requirements doc, and recording the durable lessons, which
- self-improve persists when installed; a small change set and the lessons can be handed to
- subagents from the same session. Takes the fetched file, or a PR URL a subagent fetches first.
- - **[extract-pr-review-rules](./skills/extract-pr-review-rules/SKILL.md)** — mine the conventions
- your reviewers actually enforce out of every PR review comment in a repo, triage them with you in
- one batch, and persist them through self-improve into the project's topic docs, so each rule
- loads only with its subject; later runs resume from a checkpoint and read only the new comments.
- - **[review-code-assistant](./skills/review-code-assistant/SKILL.md)** — prepare a review of a PR
- or branch with you: the diff the platform shows, the project's conventions run as a checklist,
- and candidate comments in diff order, read-only and never posted.
- - **[maintainer-review](./skills/maintainer-review/SKILL.md)** — review someone else's PR as the
- maintainer deciding whether it merges: every prior comment walked, every claim verified, then
- comment, approve or fix the contributor's branch on your go-ahead.
- [How open source projects can cope with AI slop](https://medium.com/engineering-in-the-age-of-ai/how-open-source-projects-can-cope-with-ai-slop-9ca7b015602e).
- - **[review-ticket](./skills/review-ticket/SKILL.md)** — triage a ticket or ticket set before
- anyone picks it up, saving a review with a feature walkthrough and the questions worth raising
- with the team.
- - **[verify-understanding](./skills/verify-understanding/SKILL.md)** — explain the feature back
- in your own words before building it: a teach-back conversation over its ticket review that
- probes and corrects until you are ready to implement.
- - **[check-ticket-implementation](./skills/check-ticket-implementation/SKILL.md)** — check how
- much of a ticket is already implemented in the code, saving a per-requirement status report
- (done, partial, not done, or not verifiable).
- - **[fresh-eyes-review](./skills/fresh-eyes-review/SKILL.md)** — let an agent with a fresh
- perspective review a changeset and report its findings back to the main session. [Read more about fresh-eyes-review](https://medium.com/engineering-in-the-age-of-ai/more-powerful-ai-reviews-with-fresh-eyes-bfad221748c0).
- - **[self-review](./skills/self-review/SKILL.md)** — get your changeset merge-ready before
- requesting review: a fresh-context reviewer checks it as a maintainer would, you answer every
- finding (or say `autoaccept` to let the agent decide), and a compact report for the PR proves
- the review happened. Findings that reveal a durable lesson end in an offer to run self-improve,
- when installed.
- Projects can tune the review via an optional `.agents/docs/self-review-rules.md`.
- ### Talking to humans
- Wording for what the agent writes to people, whether the user in the session or other humans.
- - **[explain-in-simple-language](./skills/explain-in-simple-language/SKILL.md)** — word
- explanations, recaps, questions to the user, and documents humans read so they are understood
- on the first read, without dumbing them down; voice stays with the caller.
- - **[use-conversational-language](./skills/use-conversational-language/SKILL.md)** — the voice for
- text that should read as if a person typed it, layered on explain-in-simple-language; used by
- the review skills for comments and replies and by rules for texts other people read and code
- comments.
- ### Code checks
- - **[run-nx-checks](./skills/run-nx-checks/SKILL.md)** — run format, lint, typecheck, test, and
- build on the affected projects of an Nx workspace and fix unambiguous failures, reporting every
- edit it made.
- ### How to install the skills
- Install all skills in one command:
- ```sh
- git clone https://github.com/eai-org/agent-toolkit.git && cd agent-toolkit && ./install.sh
- ```
- Update in one command:
- ```sh
- cd agent-toolkit && git pull && ./install.sh
- ```
- On Claude Code the install also registers a hook that runs that update for you once a day; an
- install made before it needs the command above once to pick it up
- ([docs/auto-update.md](./docs/auto-update.md)).
- Every skill is installed by default. To leave one out, name it once and the choice sticks, daily
- update included:
- ```sh
- ./install.sh --exclude use-conversational-language
- ```
- `--include <name>` puts it back. Both are repeatable.
- On Windows the skills link through junctions, so the update command above works as-is, though an
- install predating them needs one `./install.sh --force` to convert. The rules are copied instead
- and need `--force` every time (see [Windows](./docs/install-skills.md#windows)).
- How the symlink install works and the other install methods — hand-picking skills, other agents,
- [skills.sh](https://skills.sh/), the Claude Code plugin marketplace — are covered in
- [docs/install-skills.md](./docs/install-skills.md).
- ## Rules
- A set of generic, project-agnostic, opinionated rules that apply to any codebase. They are opt-in,
- installed separately from the skills.
- - **[compact-governing-docs](./rules/compact-governing-docs.md)** — run the matching compaction
- skill before writing or editing a governing doc, so it stays compact.
- - **[git-read-only-by-default](./rules/git-read-only-by-default.md)** — never commit, push, merge,
- or otherwise write to git without an explicit instruction.
- - **[no-ai-attribution](./rules/no-ai-attribution.md)** — no AI co-author trailers, "Generated with"
- footers, session links or trailers, or other AI-attribution lines on commits and PRs.
- - **[no-nonsense-comments](./rules/no-nonsense-comments.md)** — write only code comments that
- still make sense to a future reader with zero context, prefer no comment over a low-value one,
- and voice them via [use-conversational-language](./skills/use-conversational-language/SKILL.md).
- - **[plans-directory](./rules/plans-directory.md)** — save plans and similar documents under the
- project's planning directory, following a certain structure; one per repository, resolved to the
- main checkout from a linked worktree.
- - **[read-other-repos-governing-docs](./rules/read-other-repos-governing-docs.md)** — before
- editing another repo's files, read and follow that repo's governing docs — they don't auto-load.
- - **[self-contained-docs](./rules/self-contained-docs.md)** — keep planning and design docs, and
- prompts for another session, concise and executable by a fresh session with no prior context,
- writing the prompts via [prepare-prompt](./skills/prepare-prompt/SKILL.md).
- - **[self-improve-on-correction](./rules/self-improve-on-correction.md)** — when the user corrects
- something a skill or doc governs, offer to persist the lesson via
- [self-improve](./skills/self-improve/SKILL.md).
- - **[write-realistic-texts](./rules/write-realistic-texts.md)** — make text other people will
- read sound natural, and get the user's go-ahead on the wording before publishing the texts it
- names.
- - **[write-simple-explanations](./rules/write-simple-explanations.md)** — word questions,
- explanations, recaps to the user, and documents humans read via
- [explain-in-simple-language](./skills/explain-in-simple-language/SKILL.md).
- One more rule for non-English speakers ships as a copy-paste snippet rather than a file, since it
- needs the language filled in — [use-my-mothertongue](./docs/use-my-mothertongue-rule.md) makes
- your agent talk to you in your language while everything it writes into the project stays English.
- ### How to install the rules
- Because rules are always-on, they are never installed implicitly with the skills. To opt in:
- ```sh
- ./install-opinionated-rules.sh
- ```
- Auto-loaded rules are mostly a Claude Code feature — how they work, which agents support them, the
- script's options, and linking individual rules by hand are covered in
- [docs/install-rules.md](./docs/install-rules.md).
- ## Install with agentwheel
- [agentwheel](https://github.com/NestDevLab/agentwheel) installs this repo's rules **and** skills
- into your agent and keeps them in sync across Claude, Codex, Copilot, and other runtimes, from
- one source. Run it from where you want it installed (`~` for user level, or a project root):
- ```sh
- npx agentwheel install github:eai-org/agent-toolkit --adapter claude
- ```
- Other adapters, selecting individual pieces, and the OpenPack manifest are covered in
- [docs/install-with-agentwheel.md](./docs/install-with-agentwheel.md).
- ## Artifact relationships
- Solid arrows are the `requires` pairs declared in [`openpack.json`](openpack.json); dotted "when
- available" arrows are optional siblings a skill uses when present, as a next-phase hand-off or an
- in-run call; dotted "informs" arrows are rules shaping a skill's output.
- ```mermaid
- flowchart TD
- fetch_ticket["fetch-ticket"] -. when available .-> refine["refine-ticket"]
- fetch_ticket -. when available .-> attach["attach-to-ticket"]
- attach -. when available .-> refine
- fetch_pr["fetch-pr-review"] -. when available .-> refine_pr["refine-pr-review"]
- refine_pr --> refine
- refine_pr -. when available .-> plan
- refine_pr --> express["use-conversational-language"]
- review_code["review-code-assistant"] --> express
- review_code --> simple["explain-in-simple-language"]
- self_review["self-review"] --> fresh_eyes["fresh-eyes-review"]
- self_review -. when available .-> self
- simple -. when available .-> self
- harden["harden-artifact"] --> fresh_eyes
- harden -. when available .-> plan
- harden -. when available .-> manual
- harden -. when available .-> split
- harden -. when available .-> execute
- refine -. when available .-> harden
- refine -. when available .-> fetch_ticket
- refine_pr -. when available .-> fetch_pr
- refine_pr -. when available .-> harden
- plan -. when available .-> harden
- harden -. when available .-> simple
- harden -. when available .-> self
- maintainer_review["maintainer-review"] --> review_code
- maintainer_review --> fresh_eyes
- maintainer_review --> express
- realistic_rule["write-realistic-texts rule"] --> express
- nonsense_rule["no-nonsense-comments rule"] --> express
- review_ticket["review-ticket"] --> fetch_ticket
- review_ticket --> express
- review_ticket --> simple
- maintainer_review --> simple
- check_impl --> simple
- manual --> simple
- review_ticket -. when available .-> verify_understanding["verify-understanding"]
- review_ticket -. when available .-> refine
- verify_understanding -. when available .-> refine
- verify_understanding -. when available .-> review_ticket
- review_ticket -. when available .-> check_impl
- review_ticket -. when available .-> attach
- check_impl["check-ticket-implementation"] --> fetch_ticket
- refine -. when available .-> manual["create-manual-test-instructions"]
- refine -. when available .-> plan["create-implementation-plan"]
- refine -. when available .-> express
- refine -. when available .-> attach
- plan -. when available .-> split["split-plan-tasks"]
- split --> execute["execute-plan-tasks"]
- execute -. when available .-> self_review
- execute -. when available .-> plan
- execute -. when available .-> harden
- self_review -. when available .-> handover["handover"]
- handover --> express
- express --> simple
- simple_rule["write-simple-explanations rule"] --> simple["explain-in-simple-language"]
- refine -. when available .-> simple
- plan -. when available .-> simple
- refine_pr -. when available .-> simple
- verify_understanding -. when available .-> simple
- split -. when available .-> simple
- execute -. when available .-> simple
- compact -. when available .-> simple
- memory_doctor -. when available .-> simple
- agentify["agentify-project"] -. when available .-> simple
- agentify -. when available .-> self
- grill_me["grill-me"] -. when available .-> simple
- self_review -. when available .-> simple
- agentify -. when available .-> compact_docs
- agentify -. when available .-> memory_doctor
- agentify -. when available .-> context_checkup["context-checkup"]
- context_checkup -. when available .-> agentify
- context_checkup -. when available .-> memory_doctor
- self_rule["self-improve-on-correction rule"] --> self["self-improve"]
- self --> compact["compact-skill-creator"]
- self --> compact_docs["compact-docs-writer"]
- compact --> compact_docs
- memory_doctor["memory-doctor"] --> self
- extract_pr["extract-pr-review-rules"] --> self
- refine_pr -. when available .-> self
- compact_gov["compact-governing-docs rule"] --> compact
- compact_gov --> compact_docs
- plans_rule["plans-directory rule"] -. informs .-> fetch_ticket
- plans_rule -. informs .-> attach
- plans_rule -. informs .-> fetch_pr
- plans_rule -. informs .-> refine_pr
- plans_rule -. informs .-> refine
- plans_rule -. informs .-> manual
- plans_rule -. informs .-> plan
- plans_rule -. informs .-> review_ticket
- plans_rule -. informs .-> check_impl
- plans_rule -. informs .-> handover
- plans_rule -. informs .-> self_review
- plans_rule -. informs .-> extract_pr
- docs_rule["self-contained-docs rule"] --> prepare_prompt["prepare-prompt"]
- docs_rule -. informs .-> fetch_ticket
- docs_rule -. informs .-> fetch_pr
- docs_rule -. informs .-> refine_pr
- docs_rule -. informs .-> refine
- docs_rule -. informs .-> manual
- docs_rule -. informs .-> plan
- ```
- ## Contributing
- Any agent works: the repo keeps one canonical [AGENTS.md](./AGENTS.md) that every agent reads, and
- [docs/contributor-setup.md](./docs/contributor-setup.md) explains the layout and how to hook yours
- up.
Ratings and comments
Sign in to rate this skill.
Sign inFree to use. If it’s helped, pass it on.